BOLD signal response in primary visual cortex to flickering checkerboard increases with stimulus temporal frequency in older adults.

Many older adults have difficulty seeing brief visual stimuli which younger adults can easily recognize. The primary visual cortex (V1) may induce this difficulty. However, in neuroimaging studies, the V1 response change to the increase of temporal frequency of visual stimulus in older adults was unclear. Here we investigated the association between the temporal frequency of flickering stimuli and the BOLD activity within V1 in older adults, using surface-based fMRI analysis. The fMRI data from 29 healthy older participants stimulated by contrast-reversing checkerboard at temporal flicker frequencies of 2, 4, and 8 Hz were obtained. The participants also performed a useful field of view (UFOV) test. The slope coefficient of BOLD activity regarding the temporal frequency of the visual stimulus averaged within V1 regions of interest was positive and significantly different from zero. Group analysis in the V1 showed significant clusters with positive slope and no significant clusters with a negative slope. The correlation coefficient between the slope coefficient and UFOV performance was not significant. The results indicated that V1 BOLD response to a flickering visual stimulus increases as the stimulus temporal frequency increases from 2 to 8 Hz in older adults.

Introduction

Many older adults have difficulty seeing brief visual stimuli which younger adults can easily recognize [1-3]. The difficulty among older adults interferes with activities of daily living [4], including walking without falls [5] and safe driving [6-8]. Training in seeing brief stimuli improved the difficulty [9, 10]. The improvement suggests that causes of the reduced temporal-processing ability are not only attributed to optical factors, such as decreased ocular transmittance [11], increased intraocular scattering [12], and increased ocular aberration [13] in the eyes, but also to the visual processing deterioration in neuronal pathways.

One of the potential brain regions causing difficulty in the pathways may be the primary visual cortex (V1). Previous studies showed that V1 response to visual stimuli changed with age. V1 neuronal response in older monkeys was different from that of younger monkeys [14-16]. Functional neuroimaging studies in humans showed that V1 neural responses to visual stimuli in older adults were lower than those in younger adults [17-20]. These results suggest that V1 is a potential region that causes the older adults’ vulnerability to brief visual stimuli.

The difficulty of seeing brief visual stimuli in older adults is considered to be due to low neural response to brief visual stimuli. A low neural response can be considered low information processing ability in the brain. If some information processing ability in V1 becomes lower to the brief visual stimuli in older adults, V1 response (i.e., functional neuroimaging signal) to brief visual stimuli should be low. This means that the V1 response to flickering visual stimuli with higher temporal frequency is lower in older adults.

Functional neuroimaging studies of younger adults repeatedly showed that an increase of visual stimulus frequency of up to 8 Hz increases neuroimaging signal response in the V1 [21-29]. However, in older adults, the relation between the temporal frequency of the visual stimulus and the V1 response has not been clear. A positron emission tomography (PET) study by Mentis et al. demonstrated that cerebral blood flow (CBF) response to flickering stimuli increased as stimulus frequency increases in older adults [30]. An fMRI study by Cliff et al. suggested that the V1 BOLD response decreased with the increase of frequency in older adults. This result seems to be our expectation of V1 response in older adults. Another fMRI study by Fabiani et al. indicated a tendency of the increase in V1 BOLD response in older adults, although the temporal-frequency effect on BOLD response was not statistically tested [31].

A possible reason for the inconsistent results in older adults might be V1 localization methods in the studies. The past two fMRI studies did not clearly indicate that the regions in which BOLD response increased with the stimulus frequency were placed in V1. The study by Cliff et al. [32] only showed axial activation maps of group analysis to localize V1 and did not describe brain coordinates of peak voxels. The activation map indicated that the decrease was in the posterior part of V1, which means that the region with the decrease was in the peripheral visual field. Moreover, another fMRI study by Fabiani et al. [31] investigated BOLD signal averaged within V1 determined by inspection of individual anatomical images and did not show the standard brain coordinates of the region. On the other hand, the PET study by Mentis et al. [30] clearly showed a Talairach location within V1 where CBF was increased with temporal frequency. Thus, the past fMRI studies are insufficient to conclude whether increase or decrease in V1 BOLD activity as the stimulus temporal frequency increases.

Cortical surface-based analysis is superior to voxel-based analysis to investigate the V1 BOLD activity [33-37]. The V1 map, based on the cytoarchitecture of postmortem human brains, was implemented in a software suite for surface-based analysis as a standard brain map [35]. Voxel-based analysis can also use a cytoarchitectonic map of V1 [38]; however, the surface-based analysis can determine the V1 more accurately than the voxel-base analysis [37]. The reason for accuracy difference between the two analyses has been considered in that the cerebral cortical folding patterns (i.e., gyri and sulci) are effective to estimate V1 location. Additionally, surface-based analysis is more accurate for localizing the activated region and has higher efficiency for detecting activated region than voxel-based analysis [33, 34, 36]. Cortical surface is complexly folded with sulci and gyri; however, voxel-based analysis cannot possibly separate the neural activities between adjacent gyri like regions around the calcarine sulcus in V1. Thus, surface-based analysis is suitable for V1 activity studies.

This study aimed to investigate the association between the frequency of flickering stimuli and the BOLD activity within V1 in older adults, using surface-based fMRI analysis. The surface-based analysis will reveal the V1 region related to a temporal frequency change of visual stimuli more accurately than the voxel-based analysis used in previous studies [30-32]. fMRI data from 29 healthy older participants were derived from contrast-reversing checkerboard stimuli at temporal flicker frequencies of 2, 4, and 8 Hz, and the resulting BOLD within V1 was examined. Furthermore, the factors associated with the individual difference of the V1 BOLD response were investigated. We examined the correlation between individual characteristics, including visual processing speed (i.e., useful field of view [UFOV] task [39] performance), and BOLD response change with the frequency.

Materials and methods

Participants

In this fMRI experiment, 15 male and 14 female participants (mean age, 68 years; age range, 65–74 years) were included. The following inclusion criteria were assessed using a questionnaire: absence of history of eye disease with normal and corrected-to-normal vision, absence of neurological or psychiatric illness, and absence of developmental disorders; a score of ≥24 on the Mini-Mental State Examination [40]. All participants showed no pathological changes in their structural MRI images (reviewed by one experienced radiologist, NS) and were right-handed according to the Edinburgh Handedness Inventory [41].

The protocol was approved by the ethics committees of the National Institute of Physiological Sciences and Toyota Central R&D Labs., Inc. (Nagakute, Aichi, Japan). A written informed consent was obtained from all participants. This study was conducted according to the tenets of the Declaration of Helsinki.

Imaging setup

Visual stimuli were delivered using Presentation software (Neurobehavioral Systems, Berkeley, CA, USA) on a computer CF-B10CWADR (Panasonic, Osaka, Japan). An LCD projector CP-SX12000 (Hitachi, Tokyo, Japan) projected the stimuli (resolution, 1280 × 1024 pixels) on a screen in an MRI scanner room. Participants were asked to lay down in a 3 T MRI MAGNETOM Verio (Siemens, Munich, Germany) with a 32-channel head coil, and visual stimuli were observed through a mirror mounted on a head coil. The viewing distance between the eye and the screen (size, 56.8 × 42.6 cm) was approximately 180 cm. A video camera captured the eyes of the participant through the mirror. The eye images were displayed in an MRI scanning control room. A USB data acquisition unit, USB-6009 (National Instruments, Austin, TX, USA), was used to synchronize the presentation of the visual stimuli with the fMRI scan.

fMRI procedure

V1 neural activity was measured using a block-design fMRI procedure consisting of alternating rest block and stimulus block of 16 s each. In the stimulus block, a checkerboard pattern and its inverted pattern were alternately presented. The temporal flicker frequency of alternation was 2, 4, or 8 Hz, and the duty cycle of the alternation was 50%. The checkerboard pattern consisted of black and white squares (20 × 20) with a Michelson contrast of 0.97 (black: 6.77 cd/m2; white: 500 cd/m2) subtended approximately at 13.5 × 13.5°. A red circle (subtending a visual angle of 0.27° at the diameter) was overlaid on the stimulus center as a fixation point. In the rest of the blocks, the same fixation point was presented on a uniform gray background. Three runs, each consisting of six stimulation blocks and seven rest blocks with a total of 208 s duration, were performed. One of the three flicker frequencies of 2, 4, and 8 Hz was randomly assigned to each of the three consecutive stimulus blocks. The frequency order in a run was randomized for each run, which means that the frequency orders were different among participants. Participants were instructed to gaze at the fixation point throughout a run. During the checkerboard viewing task, experimenters kept watching the participants’ eyes with a monitor displaying the eyes and confirming that the eyelids were open.

MRI data acquisition

For fMRI runs, a T2*-weighted gradient-echo echo planar imaging (EPI) sequence was used. An EPI image consisted of 31 transaxial slices of 3-mm thickness each with no gaps. The slices were acquired in an ascending mode with oblique scanning. The following settings were used: repetition time (TR), 2000 ms; flip angle (FA), 76°; echo time (TE), 30 ms; field of view (FOV), 192 mm; and in-plane matrix, 64 × 64 pixels. A total of 107 EPI images were acquired for each run. A structural brain image was acquired for each participants using T1-weighted magnetization-prepared rapid gradient-echo imaging sequences with the following settings: TR, 1800 ms; TE, 2.97 ms; FA, 9°; FOV, 250 × 250 mm; in-plane matrix size, 256 × 256 pixels; slice thickness, 1 mm; 192, contiguous transaxial slices).

Visual processing speed task

The study participants performed a UFOV test before the fMRI data acquisition. A participant performed 288 test trials. In the trial, a visual stimulus was presented with a short duration selected randomly from 40, 60 100, 180, 340, and 660 ms. At the center of the visual stimulus, a randomly selected letter from E, F, H, and L was placed. In each of the four peripheral corners around the letter, a circle was placed. One randomly selected circle from the four peripheral corners was filled, and the other three were opened. After the stimulus was presented, the participant responded as to which letter was presented and which circle was filled. The correct trials were that both responses to the center and the peripheral stimulus were correct. The UFOV test performance of the participant was stimulus duration of 53% correct rate of the responses estimated from all trial responses of the participant. To estimate the duration, a logistic-regression model was fitted to the UFOV trial response with respect to the stimulus duration. Using the model, the stimulus duration of 53% correct rate of the response was estimated; thus, shorter stimulus duration represents higher UFOV performance. The detailed task and analysis procedures have been described elsewhere [39].

fMRI data analysis

We conducted two fMRI analyses: categorical analysis which shows BOLD activity against the resting baseline for each stimulus frequency (i.e., 2, 4, and 8 Hz) and parametric analysis which shows the BOLD activity change regarding the stimulus frequency increase (i.e., slope) and the averaged BOLD response to the stimulus block regardless of its stimulus frequency (i.e., offset). We performed both analyses using the FreeSurfer software suite (version 7.7.1) [42] on Linux (CentOS 7). The first step of preprocessing was common to the two analyses.

Preprocessing

Brain surfaces were segmented from T1 images of each participant. If the dura mater were included in the surface image, we edited the skull-stripped images (i.e., brainmask.nii.gz) manually to exclude the dura mater, and the surface images were corrected. The brain surfaces of the left and right hemispheres, derived from T1 images of each participant, were tessellated with vertices and were inflated without stretching. The inflated surfaces were projected on a standard sphere of the brain using information of the gyri and sulci. Based on the mapping information between the standard sphere and individual sphere, individual V1 region was decided using the cytoarchitectonic V1 map (i.e., vcAtlas) [35].

Brain activity was analyzed using FsFast (FreeSurfer Functional Analysis Stream) tools. The first three EPI images of each run were discarded to allow for stabilization of the magnetic field. All the remaining EPI images were corrected for slice timing and head motion. The corrected EPI images were registered on the surfaces of individual and of standard averaged brain (i.e., fsaverage). EPI signal mapped on the surfaces were smoothed using a two-dimensional Gaussian kernel with a full width at half-maximum of 5 mm.

Categorical fMRI analysis of visual stimulus frequency

Individual analysis

For individual level analysis, the expected BOLD activity for each run and for each flicker frequency was modeled as a regressor of interest using the boxcar function, convolved with a hemodynamic response function (HRF), which was SPM canonical without derivatives. The regressors were fitted to the EPI time series for each vertex on individual and standard averaged surface. Estimates obtained by the fit were considered to represent the BOLD activity for each flicker frequency.

Regions of interest (ROI) analysis

To analyze the overall V1 activity, we conducted most ROI analyses using Matlab 2019a with Statistics and Machine Learning Toolbox (Mathworks Inc.). Mauchly’s test of sphericity and one-way repeated measures analysis of variance (ANOVA) were conducted using anovakun 4.8.5 software package [43] on R 4.0.5. Vertices within the ROIs satisfied the following two conditions: 1) significantly activated by the stimulus with surface-based analysis (uncorrected P < 0.05) and 2) within V1 defined by vcAtlas for each participant. The magnetic resonance (MR) signal changes were averaged across the ROI for each frequency and each participant (S1 File). After Mauchly’s test of sphericity, repeated-measured one-way ANOVA, with frequency as the within-subjects factor, was applied to the averaged MR signal changes. Tukey’s tests were performed as post-hoc tests of the ANOVA. Statistical significances of the sphericity test, the ANOVA and the post-hoc tests were set at P < 0.05. To confirm the increase of the MR signal change within the ROI with the frequency, Pearson’s product-moment correlation coefficients between the frequency and MR signal change within the ROI were tested for their difference from zero by one-sample t-test. The significance level of the t-test was P < 0.05.

Group analysis

To compare the brain activity among all participants between the frequencies, group level analysis of the brain activity on an averaged brain surface was conducted using permutation tests using FsFast tools. The following three contrasts were tested: 1) brain activity at 8 Hz was higher than that at 2 Hz (8 Hz > 2 Hz), 2) brain activity at 4 Hz was higher than that at 2 Hz (4 Hz > 2 Hz), and 3) brain activity at 2 Hz was higher than that at 8 Hz (2 Hz > 8 Hz). Significance threshold of the test was set at cluster level corrected P < 0.05.

Parametric fMRI analysis

The slope and offset contrasts for each run were modeled as two regressors. The slope regressor was a boxcar function whose amplitude was modulated by the stimulus frequency. The offset regressor was a boxcar function whose stimulus onset and offset corresponded to those of stimulus blocks, respectively, with constant amplitude regardless of the frequency. Both regressors were convolved with a HRF, SPM canonical without derivatives. The regressors were fitted to the EPI time series for each vertex on individual and standard averaged surface.

To analyze the overall slope within V1, we conducted ROI analyses using Matlab 2019a with Statistics and Machine Learning Toolbox (Mathworks Inc.). Vertices within the ROIs satisfied the following two conditions: 1) the offset contrast was significant and 2) within V1 defined by vcAtlas for each participant. We averaged the effect size of the slope across the ROI for each participant (S2 File). The averaged effect size of the slope and the offset within the ROI were tested for the difference from zero across the participants using a one-sample t-test. The significance level was set at P < 0.05.

To investigate the individual difference of the V1 BOLD response to the visual stimulus frequency, we tested Pearson’s product-moment correlation coefficient between the effect size of the slope or the offset within the ROI and the individual characteristics of age, MMSE, or UFOV performance. The significance level was set at P < 0.05 for the correlation test.

A group level analysis of the positive and negative contrasts of the slope and the offset on an averaged brain surface was conducted using permutation tests with FsFast tools. The significance threshold of the test was set at cluster level corrected P < 0.05.

Results

This study first explored MR signal change within V1 for each participant. Fig 1 shows the highest peak MR signals in the left and right V1 among the three representative subjects. The peaks were found in and around the calcarine sulcus within V1. The MR signal on the peaks increased with the temporal frequency for each hemisphere and each participant. Among all 29 participants, 13 participants showed that peak MR signal increase with the stimulus frequency (i.e., the order of MR signal change is 8 Hz > 4 Hz > 2 Hz, which is identical of Fig 1) in V1 of both hemispheres, and three participants showed that the peak MR signal increase in V1 of one hemisphere.

Fig 1

Peak BOLD activities in V1 increase as temporal flicker frequency increases (2, 4, and 8 Hz) in three representative participants.

(A) MR signal changes are overlaid on an inflated brain surface of each participant in the left occipital region. The brain surfaces are shown from a posterior/medial viewpoint. White lines on the brain surfaces indicate the border of V1, based on cytoarchitecture using vcAtlas [35]. Dark and light gray regions represent sulci and gyri, respectively. Calcarine sulci are dark gray regions passing near the center of V1 and extending from anterior to posterior within the region bordered by the white lines (i.e., V1). Peaks of MR signal change within the region bounded by the white lines (i.e., V1) increase from the left (i.e., 2 Hz) to the right (i.e., 8 Hz). (B) Results in the right occipital region correspond to (A).

ROI analysis

The association between MR signal on V1 and the temporal frequency in the ROI was examined. The MR signal averaged in V1 increased with the temporal flicker frequency (Fig 2 and Table 1). A repeated-measured one-way ANOVA was performed to test the effect of the frequency on the MR signal in the ROI. Mauchly’s test of sphericity was accepted (χ2 (2) = 4.53; P = 0.10). The ANOVA revealed a significant effect of the temporal frequency on the MR signal averaged across the ROI (F(2,56) = 15.1; P = 5.52 × 10−6, ηP2 = 0.35). A post-hoc analysis using Tukey’s test indicated that the MR signal changes of 4 Hz was larger than that of 2 Hz (P = 0.0012, 95% CI = 0.05–0.22) and that of 8 Hz was larger than that of 2 Hz (P = 0.00017, 95% CI = 0.11–0.34). However, the MR signal changes of 8 Hz was not significantly larger than that of 4 Hz (P = 0.082, 95% CI = −0.01–0.19). Additionally, to confirm the increase of the MR signal in the ROI with the temporal frequency, one-sample t-test of the mean correlation coefficients between the frequencies and mean MR signal averaged across the ROI was performed. The mean correlation coefficient (mean R = 0.63, SD = 0.56) was significantly larger than zero (one-sample t-test, t(28) = 6.07, P = 1.50 × 10−6, 95% CI = 0.42–0.85). These results showed that the V1 BOLD activity increased with the temporal flicker frequency.

Fig 2

Magnetic resonance (MR) signal change in V1 increases with temporal flicker frequency.

The MR signal change was averaged across V1 regions of interest (ROIs) for each participant and across 29 participants. The V1 ROIs were cortical vertices activated significantly (vertex-level uncorrected P < 0.001) within the V1 based on cytoarchitectonic atlas. Error bars represent the standard errors of the MR signal change. The main effect of the flicker frequency in one-way analysis of variance was found to be significant (P = 6.5 × 10−6). **: P < 0.01 (Tukey’s test).

Table 1

Magnetic resonance (MR) signal change averaged within each region of interest and across all participants (n = 29).

Temporal flicker frequency (Hz)	MR signal change (%) Mean (SD)
2	2.53 (0.77)
4	2.66 (0.78)
8	2.75 (0.75)

SD: standard deviation.

Group analysis was conducted to validate that V1 activity increases with the temporal frequency. The stimulus with each temporal frequency significantly induced V1 brain activity on both hemispheres (cluster-wise corrected P < 0.05) (Fig 3A and 3B, and Table 2). Furthermore, similar relations obtained by the post-hoc tests of the ROI analysis were confirmed. V1 activity at 8 Hz was significantly higher than that at 2 Hz (cluster-wise corrected P < 0.05) (Fig 3C and Table 2). V1 activity at 4 Hz was also significantly higher than that at 2 Hz (cluster-wise corrected P < 0.05) (Fig 3D and Table 2).

Fig 3

Significant brain activity derived from group analysis (n = 29).

Red regions indicate statistically significant clusters activated by each temporal flicker frequency (2, 4, and 8 Hz) on the left (A) and right (B) hemispheres. (C) Statistically significant clusters in which the brain activity of 8 Hz is higher than that of 2 Hz (8 Hz > 2 Hz contrast). (D) Statistically significant clusters in which the brain activity of 4 Hz is higher than that of 2 Hz (4 Hz > 2 Hz contrast). Brain surfaces are shown from a posterior/medial viewpoint. White lines on the brain surfaces indicate the border of V1. Statistical significance is cluster-wise corrected P < 0.05.

Table 2

Statistically significant clusters for each contrast by the group analysis (n = 29).

Contrast	Side	Cluster			Maximum peak vertex
					Vertex-level uncorrected P	MNI coordinates
		Cluster-wise corrected P	(mm2)	Cortical region		x	y	z
2 Hz	Left	0.003	6132.02	Lateral occipital	3.3 × 10−18	-12.3	-102.0	3.6
	Right	0.003	7160.79	Lateral occipital	3.5 × 10−16	14.9	-101.3	5.9
4 Hz	Left	0.003	5097.11	Lateral occipital	6.9 × 10−18	-13.7	-101.7	3.9
	Right	0.003	6037.11	Lateral occipital	9.8 × 10−16	17.0	-100.9	3.7
8 Hz	Left	0.003	6626.71	Lingual	3.0 × 10−18	-9.0	-92.6	-10.0
	Right	0.003	7322.04	Lateral occipital	4.9 × 10−16	15.4	-101.0	6.0
8 Hz > 2 Hz	Left	0.003	3724.19	Lateral occipital	4.2 × 10−10	-20.6	-97.4	6.5
	Right	0.003	3707.5	Lateral occipital	6.7 × 10−10	14.1	-99.0	11.6
4 Hz > 2 Hz	Left	0.003	885.56	Lateral occipital	1.1 × 10−6	-14.6	-101.0	4.7
	Right	0.003	968.41	Lateral occipital	1.4 × 10−7	19.1	-97.3	14.9

MNI: Montreal Neurological Institute.

Note: Cortical regions are indicated by Desikan-Killiany atlas [44].

To find that V1 region decreased with the temporal frequency, the contrast between 2 Hz > 8 Hz was tested. This study could not find any V1 region in which activity at 2 Hz was larger than that at 8 Hz.

Finally, this study checked that the peak positions of the individual analysis of the three representative subjects shown in Fig 1 were on the group analysis results of the 8 Hz > 2 Hz contrast in Fig 3C. Fig 4 shows that every peak shown in Fig 1 were in regions overlapping between V1 and the region of 8 Hz > 2 Hz.

Fig 4

Individual peak responses to 8 Hz are within the V1 region in which the brain activity of 8 Hz was higher than that of 2 Hz in the group analysis.

Peak brain responses to 8 Hz among the three representative subjects in Fig 1 are overlaid on the group analysis results of 8 Hz > 2 Hz contrast in Fig 3C. Blue and green lines on the brain surfaces indicate the border of the 8 Hz > 2 Hz region. White lines on the brain surfaces indicate the border of V1.

This study has examined the slope and the offset of the BOLD activity within V1 for each participant. Fig 5 shows the highest peak of the slope and the offset in the left and right V1 among the three representative subjects. The peaks were found in and around the calcarine sulcus within V1.

Fig 5

Peak slope and offset in V1 among the three representative participants.

(A) Peak effect sizes of slope and offset contrasts are overlaid on an inflated brain surface of each participant in the left occipital region. The brain surfaces are shown from a posterior/medial viewpoint. White lines on the brain surfaces indicate the border of V1, based on cytoarchitecture using vcAtlas [35]. Dark and light gray regions represent sulci and gyri, respectively. Peaks of the slope and the offset within the region bounded by the white lines (i.e., V1). (B) Results in the right occipital region correspond to (A).

This study confirmed the increase of the BOLD activity in the ROI with the temporal frequency using the slope coefficient. In one participant, any vertex, which met the ROI inclusion criteria, was not found. Thus, we performed the analysis with 28 participants. The BOLD slope and offset coefficients averaged within the ROI were tested across the 28 participants using a one-sample t-test. The mean slope coefficient (mean coefficient = 0.049, SD = 0.56), and the mean offset coefficient (mean coefficient = 2.45, SD = 0.78) were significantly larger than zero (one-sample t-test, t(27) = 5.94, P = 2.38 × 10−6, 95% CI = 0.032–0.065, and t(27) = 16.5, P = 1.27 × 10−15, 95% CI = 2.14–2.75, respectively). The positive slope coefficient showed that the V1 BOLD activity increased with the temporal flicker frequency.

This study confirmed the brain areas in which the BOLD activity increased with the slope contrast of the group analysis. In the positive slope contrast, significant regions were found in the posterior part in the V1 on both hemispheres (cluster-wise corrected P < 0.05) (Fig 6A, and Table 3). In the negative slope contrast, there were no significant clusters in the V1 on both hemispheres.

Fig 6

Significant regions of slope and offset by group analysis (n = 29).

The significant slope (A) and offset clusters (B) were overlaid on the standard inflated brain surface. Red and blue regions indicate positive and negative contrast, respectively. Brain surfaces are shown from a posterior/medial viewpoint. White lines on the brain surfaces indicate the border of V1. Statistical significance level was set at cluster-wise corrected P < 0.05.

Table 3

Summary of statistically significant clusters for the slope and offset contrasts by group analysis (n = 29).

Contrast	Side	Cluster			Maximum peak vertex
					Vertex-level uncorrected P	MNI coordinates
		Cluster-wise corrected P	(mm2)	Cortical region		x	y	z
Positive slope	Left	0.003	3840.12	Lateral occipital	1.0 × 10−9	-20.9	-97.0	6.5
	Right	0.003	3922.04	Pericalcarine	3.5 × 10−9	12.1	-93.6	8.4
Positive offset	Left	0.003	4938.82	Lateral occipital	1.7 × 10−16	-15.3	-100.8	4.5
	Right	0.003	6488.09	Lateral occipital	1.2 × 10−15	16.1	-100.5	7.0
Negative offset	Left	0.003	3281.61	Cuneus	2.9 × 10−10	-5.6	-86.2	19.2
	Right	0.003	5168.37	Cuneus	2.9 × 10−10	7.1	-81.5	16.7

MNI: Montreal Neurological Institute.

Note: Cortical regions are indicated by Desikan–Killiany atlas [44]. Significant clusters with negative slope were not found.

Additionally, the study examined the positive and negative offset contrast. The significant positive and negative areas were on the posterior and the anterior part in the V1 on both hemispheres, respectively (Fig 6B and Table 3).

Effect of individual characteristics on BOLD response

To investigate the individual difference of V1 BOLD response to the stimulus frequency, we examined the correlation between the BOLD slope or offset coefficients obtained by the parametric ROI analysis and the individual characteristics of age, MMSE, or UFOV test performance (Table 4). All of the correlations were not significant.

Table 4

Correlation coefficients between BOLD responses by parametric analysis and individual characteristics (n = 28).

Variables	Mean (SD)	Slope		Offset
		R (95% CI)	P	R (95% CI)	P
Age	68.4 (3.0)	0.09 (-0.29–0.45)	0.63	0.06 (-0.42–0.32)	0.77
MMSE	28.4 (1.7)	0.03 (-0.34–0.40)	0.87	0.34 (-0.04–0.63)	0.08
UFOV	336.0 (260.0)	-0.16 (-0.50–0.23)	0.41	0.01 (-0.37–0.38)	0.97

SD: standard deviation, R: Pearson’s product-moment correlation coefficients, CI: confidence interval, MMSE: Mini-Mental State Examination, UFOV: useful field of view.

Note: MMSE represents the total score of MMSE test, while UFOV represents UFOV test performance which is visual stimulus duration in ms when the correct response rate is 53%.

Discussion

This study has investigated how the temporal frequency of flickering checkerboard stimuli modulate V1 BOLD activity in older adults, using surface-based analysis. The categorical analysis showed that the BOLD activity increases with flicker frequency from 2 to 8 Hz within the V1 region activated by visual stimuli. The one-way ANOVA of BOLD activity in the V1 ROI showed a statistically significant main effect in flicker frequency. Post-hoc ANOVA test showed that BOLD activity of 8 Hz in the ROI was higher than that of 2 Hz and that BOLD activity of 4 Hz was higher than that of 2 Hz. The correlation coefficients between the temporal frequency and V1 BOLD activity were found to be significantly positive. Group analysis showed similar results of the post-hoc test; significant BOLD activity of 8 Hz > 2 Hz and 4 Hz > 2 Hz contrasts were found in V1. Conversely, any regions in which BOLD activity of 2 Hz was significantly higher than that of 8 Hz could not be found in V1. The parametric analysis showed the same relation derived by the categorical analysis. Slope coefficient averaged in the V1 ROI was positive and statistically different from zero. Group analysis showed significant clusters with a positive slope and no significant clusters with negative slope, in the V1.

This study is considered to locate BOLD response and V1 more precisely than the past studies using voxel-based analysis [30-32]. Surface-based analysis possibly located the region in which BOLD activity increased with the temporal frequency more precisely than those of the past functional imaging studies which used voxel-based analysis [30-32]. The surface-based analysis can separate the brain activity around folded brain surfaces [33, 34, 36]. Furthermore, V1 in this study is located more precisely than in the past fMRI studies [31, 32]. This study used a template atlas based on cytoarchitecture of postmortem brains (i.e., vcAtlas) to localize V1. The atlas well matched the V1 which was retinotopically defined [35]. Inversely, the past fMRI studies [31, 32] defined V1 based on human observation.

Our results are consistent with the PET study by Mentis et al. [30]. The study demonstrated that V1 activity significantly increased with temporal stimulus frequency in older adults. The location in which Mentis et al. found the increase in V1 activity was close to the peak position of MR signal in our results of the individual and the group analysis. The locations of both PET study and this study were around the posterior part of the calcarine sulcus. These regions were near the center of the visual field, in which flickering visual stimuli were presented.

Our results are inconsistent with the fMRI study by Cliff et al. [32], who suggested that the V1 brain region in which the response of BOLD fMRI activity to flickering checkerboard stimuli decreased with its frequency in older adults [32]. We could not find any significant V1 region in which the BOLD signal decreased with flicker frequency from the comparison between the BOLD activity at 2 and 8 Hz.

The results of this study suggest that the difference between the V1 activity changes with the temporal frequency of younger and older adults may be low. This study showed that the V1 activity increased with the temporal frequency from 2 to 8 Hz in older adults. Previous studies among younger adults [21-29] also showed the increase in V1 brain response with stimulus temporal frequency up to 8 Hz. According to the present and the previous results, it is expected that the temporal-frequency dependence of V1 activity up to 8 Hz may be similar between younger and older adults.

Negative BOLD activities in V1 were found in the offset contrast by the parametric group analysis (Fig 6B). The regions with negative BOLD activities were adjacent to the regions with positive BOLD activities within V1. Past human fMRI studies repeatedly showed the same pattern of V1 BOLD activity [45-47]. In addition, monkey electrophysiological studies confirmed this spatial distribution of neuronal activity in V1 [48]. The negative BOLD activity of the offset contrast in this study is consistent with the past studies [45-48]. These facts support that our experiment and analysis were appropriate.

This study has two limitations. In order to match the study design to the previous studies [30-32], this study used a passive viewing task which did not require participants to perform a vigilance task during flicker stimulus viewing. The passive viewing task might induce sleepiness in participants and might affect V1 BOLD response. A recent study about BOLD response change with temporal frequency of flickering visual stimulus in younger adults used vigilance task during stimulus viewing and excluded participants who did not meet the criteria of vigilance task performance [21]. Although the task in this study was passive, the experimenter monitored that the eyelids of the participants were open during the task. Thus, this limitation may be a potential concern in our study. Future studies are required to use a vigilance task and to include participants who meet the criteria of vigilance task performance. The other limitation is that fMRI analysis used the canonical HRF without derivatives. This assumes that the HRF does not change with age. However, the HRF is changed with age [49] and unfitted HRF decreases the statistical power and the effect size of BOLD signal change [50]. Use of appropriate HRF according to older adults may derive higher BOLD activity than this study.

The correlation coefficient between the BOLD slope and the UFOV test performance was negative, although the correlation coefficient was not significant (Table 4). This negative correlation means that the participants with the lower UFOV test performance have a lower BOLD slope coefficient in V1. This negative correlation tendency corresponds to the expectation that the participants who are vulnerable to brief visual stimuli have low V1 response to the higher frequency. The improvement of the vigilance control during the task and using an appropriate HRF for data analysis, as discussed previously, might help to get a clearer relation.

Contrary to our expectation that V1 BOLD activity in older adults decreases when the visual stimulus frequency increases, all results of the group analysis showed no V1 region wherein the BOLD response decreased with frequency. The results of the categorical analysis showed that no V1 region wherein the 2 Hz BOLD activity were higher than the 8 Hz BOLD activity and the correlation coefficients between the frequency and the BOLD signal at the V1 ROI were significantly positive. Additionally, the parametric analysis results showed no V1 regions with negative slope coefficients. However, the correlation coefficient between the slope and UFOV test performance is negative, although no statistically significant difference was observed. This means that the participants with lower slope tended to have lower UFOV performance. This was consistent with our expectation. The above results present that the slope is positive for the older adults as a whole, however, individual differences in the slope among the older adults may be large and lower slope may impair the visual processing speed.

Conclusion

In older adults, the V1 BOLD response to flickering visual stimulus increases as the temporal flicker frequency increases from 2 to 8 Hz.

MR signal change in V1 ROIs.

MR signal change averaged across the V1 ROI for each participant (i.e., mri_segstats command output of hOc1). Table 1 was derived from this data.

(XLSX)

Click here for additional data file.

Effect size of the slope, and the offset in V1 ROIs, and data of individual characteristics.

The slope and offset effect size across the V1 ROI for each participant (i.e., mri_segstats command output of hOc1) and individual charateristics of age, MMSE, and UFOV performance. ROI analysis results of the parametric analysis were derived from this data.

(XLSX)

Click here for additional data file.

NIfTI image files of group analysis.

Statistically significant clusters of BOLD contrasts which are 2 Hz, 4 Hz, 8 Hz, 8 Hz > 2 Hz, and 4 Hz > 2 Hz in Fig 3 and positive slope and positive and negative offset in Fig 6. These files can be mapped on the averaged brain surface (i.e., fsaverage) using a visualization tool (i.e., Freeview) in FreeSurfer.

(ZIP)

Click here for additional data file.

26 Mar 2021

PONE-D-21-00897

BOLD signal response in primary visual cortex to flickering checkerboard increases with stimulus temporal frequency in older adults

PLOS ONE

Dear Dr. Uchiyama,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration of the reviews by two experts in the field, I feel that, while your paper has considerable merit, it does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, I would like to invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The first reviewer (R1) found that the rationale for the study was presented clearly but suggested that clarification with respect to how the methods of the current study could resolve the inconsistent findings you highlight in your review of the literature. R1 also posed some questions, specifically regarding how you assured that participants maintained attention on the stimuli (R2 also raised this point), ROI validity, your modeling and individual differences that are apparent in your results.  Finally R1 suggested some additional considerations for your interpretation of results.  These considerations, along with useful suggestions R1 made regarding additional analyses should be carefully considered in your revision.

The second reviewer (R2) pointed at some inaccuracy in how you described your result of principal interest (Cliff et al., 2013).  R2 encouraged more careful framing of the problem at hand based on a carefully considered description of the Cliff results.  In this same vein, R2 was concerned about your interpretation of results insofar as your study does not feature age-differential comparisons.  You must be sure to be careful in characterizing your results with respect to age comparisons when none in fact have been made.  Like R1, R2 asked for more detail regarding the descriptions of your analyses.  Finally, R2 also pointed out variance between your results-reporting and the PLOS ONE SAMPL guidelines.

It is clear that the reviewers carefully read and constructively criticized your work.  I strongly encourage you to revise your manuscript along the lines they have provided.

Please submit your revised manuscript by May 10 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

I look forward to receiving your revised manuscript.

Best regards,

Bart Rypma

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please note that PLOS ONE does not copy edit accepted manuscripts (https://journals.plos.org/plosone/s/criteria-for-publication#loc-5). To that effect, please ensure that your submission is free of typos and grammatical errors.

*In line with PLOS' guidelines on detailed reporting (https://journals.plos.org/plosone/s/criteria-for-publication#loc-3), please ensure that you have provided sufficient detail on participant recruitment in the Methods section.

3.  Thank you for stating the following financial disclosure:

"Toyota Central R&D Labs., Inc. (TCRDL) received annual research funding from its

stockholder companies (Toyota Industries Corporation, Toyota Motor Corporation,

Aichi Steel Corporation, JTEKT Corporation, Toyota Auto Body Co., Ltd., Toyota

Tsusho Corporation, Aisin Seiki Co., Ltd., Denso Corporation, Toyota Boshoku

Corporation). This study was performed as part of the self-motivated research activities

in TCRDL. Thus, the funders had no role in the study design, data collection and

analysis, and the decision to publish."

We note that one or more of the authors is affiliated with the funding organization, indicating the funder may have had some role in the design, data collection, analysis or preparation of your manuscript for publication; in other words, the funder played an indirect role through the participation of the co-authors. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please do the following:

a) Review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. These amendments should be made in the online form.

b) Confirm in your cover letter that you agree with the following statement, and we will change the online submission form on your behalf:

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

4 Thank you for stating the following in the Competing Interests section:

"I have read the journal’s policy, and the authors of this manuscript have the following

competing interests: YU, HS, TA, and AT were employed by Toyota Central R&D

Labs., Inc. The authors have not applied for any patent related to this study."

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This study is an analysis of BOLD response in primary visual cortex during basic visual processing in a sample of older adults. Prior literature is mixed on whether or not older adults show increased BOLD response in response to increasing visual frequency rates, and the authors aim to test this by using surface-based analyses which may be more sensitive to anatomical specifics of primary visual cortex. The paper is well written and the analyses straightforward. My comments primarily regard clarification on some of the methods/analysis and suggestions to explore potential individual difference effects.

1. Introduction: The authors do a good job outlining the inconsistent findings from previous studies and possible reasons for the inconsistencies. I think the last paragraph of the Intro would also benefit from more explicit statements about how their study design/methods (e.g., surface based analyses) can help to resolve prior inconsistent findings.

2. Methods: In the discussion the authors mention the task was completely passive viewing, but were any study procedures implemented to ensure that participants remained awake/had their eyes open during the checkerboard task? A few examples of what I mean: eye-tracking, or a camera positions on the participant’s eyes that the experimenter could monitor, or a post-scan debrief asking if the participant remained awake?

3. Methods: FreeSurfer’s segmentation of the cortical surface can sometimes be problematic, especially at the back of the head where dura is frequently included within the surface boundary of posterior occipital regions. Given the importance of this surfaced-based analysis, where the FreeSurfer surfaces checked or edited to ensure no dura matter was included in the primary regions of interest?

4. Methods: For the first-level individual level analysis, why did the authors choose to model each frequency separately (i.e., three regressors) rather than having one parametric regressor that modeled the different levels of frequency (2,4,8 as weights)? A first-level parametric regressor may be more appropriate if the authors anticipate that the change in BOLD response in response to change in frequency was a dose-response type effect (e.g., BOLD showed a greater increase from 4hz to 8hz compared to 2hz to 4hz).

5. Results: Although there is a group effect of increasing frequency being associated with increased BOLD response, only 16 out of 29 participants showed this pattern at the individual level (for at least one hemisphere). Where there any characteristics of the participants that did not show this effect that might explain this (ie., Were they the oldest participants? did they have worse vision? Were they less attentive during scanning?). It would be quite interesting if the authors could quantify the amount of BOLD increase in response to frequency (perhaps through a first-level parametric regressor as suggested above) and link it to individual difference measures (e.g., visual acuity, visual speed of processing).

6. Discussion: Given that the authors find this increase in BOLD in response to increasing frequency in older adults (as is found in younger adults), might this suggest that the difficulties older adults have with visual stimuli are occurring further along the visual processing stream in higher level visual areas? Or perhaps top-down processes are interfering with visual processing? As stated above, given the somewhat large individual differences in how many older adults showed this specific pattern of frequency related BOLD increase, it would be interesting to see if this was predictive of some kind of visual processing behavioral outcome (which might help understand if visual deficits are happening further down the processing stream).

Reviewer #2: The article by Uchiyama and colleagues submitted for consideration to PLOS One examined BOLD signal changes in older (but not younger) adults in visual cortex in response to flickering checkerboard stimuli, the frequencies of which were parametrically varied between 2, 4, and 8 Hz. Results indicated that, contrary to results reported in Cliff et al. (2013), BOLD signal in fact increases with increasing flicker frequency in older adults, in similar fashion to the pattern widely observed in younger adults, and found no part of visual cortex where BOLD signal at 2 Hz was higher than at 8 Hz.

Overall, the article is well-written, though it may benefit from language editing in a few select places (Enago did a reasonably good job, but it can be improved further). I do have several concerns that I believe, if addressed, will significantly strengthen the manuscript.

▪ The central premise of the article seems to be motivated by a single 2013 study by Cliff et al. that showed "decreasing BOLD signal" with increasing flicker frequency. Because it plays so prominently in the rationale behind the study's conception, I would request that the reasoning be more thoroughly fleshed out.

▫ The Cliff et al. (2013) study does not actually show decreasing BOLD signal in older adults with increasing checkerboard flicker frequency, despite how Figure 4 from that article frames the result. The actual quantity depicted is something called the SSQRatio, which is a ratio of sums of squares, and thus does not account for degrees of freedom. This makes interpretation of any differences in the quantity much less straightforward than those of a traditional/parametric test statistic, and in fact, the authors themselves refer to the statistic itself as a goodness-of-fit statistic. Additionally, while the Cliff et al. article does mention that a significant interaction effect between age group and presentation rate was observed for the SSQRatio in bilateral primary visual cortex, it curiously does not provide any statistical metadata regarding that result.

▫ Other concerns with how the authors from the Cliff et al. article processed the imaging data seem to be sufficient to me to question the result (taken together with the first and second points above). Rather than substantially rewriting the discussion or spending unwarranted manuscript real estate to denigrate another article, I would encourage the authors to reframe the introduction to be less dependent upon that article in motivating the study. I believe the authors of the present article got the treatment of Cliff et al. right in the Discussion section, where it is far less featured.

▪ I am also curious as to the authors' choice of sample for the study. The interest was very obviously in potential effects of age on the BOLD signal in response to visual stimuli, but knowing that the Cliff et al. study could serve as a (however imperfect) model, why were younger participants not sampled as well? This is not a fatal flaw in design; far from it. But as no data for younger adults are presented here, I would ask the authors to either remove or significantly reword the last sentence from the abstract, as without a younger group, the question of how age affects V1 response is something this study cannot address. The Discussion paragraph that speaks to this (lines 287-292) begins in more appropriately worded fashion, although the last sentence should probably be softened as well, perhaps even by changing "almost the same" to "similar".

▪ The Materials and Methods section mentions convolution of a boxcar function with a hemodynamic response function. I would request the authors to be more specific here. Which one? Gamma? Double-gamma (similar to the SPM "canonical"? The choice of HRF is very important, as pointed out in Lindquist et al. (2009), especially knowing that older adults were scanned. See West et al. (2019) for concerns regarding changes with age to the BOLD-HRF.

▪ The Materials and Methods section mentions that FreeSurfer/FsFast was used for neuroimaging data analysis, but I was not clear whether this was used for post-hoc statistical testing as well. I would request the authors to be more specific here.

▪ Additional opportunities for language editing did appear from time to time as I read through this manuscript. For example, the first sentence of the ROI analysis subheading of the Materials and Methods section (line 158), sounds more correct as: "To analyze overall V1 activity, ROI analyses were conducted." As I said above, minor changes, but they could be impactful nonetheless.

▪ The authors' reporting of statistical test outcomes in the Results section does not conform to the "SAMPL" guidelines listed in PLOS One's author guidance. Per those SAMPL guidelines, "Avoid relying solely on statistical hypothesis testing, such as P values, which fail to convey important information about effect size... P values are not sufficient for re-analysis. Needed instead are descriptive statistics for the variable being compared, including sample size of the groups involved, the estimate (or “effect size”) associated with the P value, and a measure of precision for the estimate, usually a 95% confidence interval."

▫ In reading through the Results section, the authors have indeed provided only p-values. I would request that they provide estimates of effect size and, as their chosen alpha was 0.05, 95% confidence intervals.

▫ Per the SAMPL guidelines, p-values should not be reported as inequalities. It seems to me as though this is done throughout as a reminder when a significant result is disclosed that it is, indeed, significant, but this is unnecessary and potentially confusing. I would request they remove mentions of "< 0.05".

▫ The SAMPL guidelines are perfectly clear as to the use of "NS": it is to be avoided. As the authors provided the actual p-value in the results (p = 0.054), I would request they simply remove the "NS" from the figure and its mention from the caption.

▪ As to the limitation mentioned immediately prior to the Conclusion (lines 293-300), were participants monitored during scanning? If so, this could be mentioned as a point to help offset the potential concern of this limitation for readers.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Jenny Rieck

Reviewer #2: Yes: Monroe P Turner

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

19 May 2021

Please find the attached file labeled "Response To Reviewers".

Submitted filename: Response_To_Reviewers20210518.docx

Click here for additional data file.

26 Aug 2021

PONE-D-21-00897R1

BOLD signal response in primary visual cortex to flickering checkerboard increases with stimulus temporal frequency in older adults

PLOS ONE

Dear Dr. Uchiyama,

Thank you for submitting your manuscript to PLOS ONE. The paper was re-evaluated by two experts on the field, both of them found that the manuscript improved a lot, however, there are several important concerns that still should be addressed.Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by 24th of October, 2021. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Andrea Antal, PhD

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: The resubmitted article by Uchiyama and colleagues examined BOLD signal changes in older (but not younger) adults in visual cortex in response to flickering checkerboard stimuli, the frequencies of which were parametrically varied between 2, 4, and 8 Hz. Results indicated that, contrary to results reported in Cliff et al. (2013), BOLD signal in fact increases with increasing flicker frequency in older adults, in similar fashion to the pattern widely observed in younger adults, and found no part of visual cortex where BOLD signal at 2 Hz was higher than at 8 Hz.

The authors, by and large, did an excellent job in responding to the concerns I raised in my initial review: the study is more appropriately motivated without undue reliance on the Cliff et al. result, methods were made more detailed, and results were brought in line with SAMPL guidelines. A few manageable but very important concerns remain, however, and I believe the manuscript will be acceptable for publication if these can be addressed.

• Regarding the choice of the HRF (previous comment 2-4), one critique that remains is the use of a canonical HRF in older populations. Lindquist et al. (2009) elegantly make the point that the choice of HRF used to model measured BOLD signal has important consequences for results and their interpretation. The authors chose the canonical HRF from SPM, a double-gamma function widely employed in the analysis of fMRI data. However, as pointed out in West et al. (2019), this function makes rigid assumptions regarding evolution of the signal’s time-course (e.g., the canonical HRF’s time-to-peak is not free to vary, despite the finding in that study that it is significantly greater in older than in younger adults). This is especially true given that temporal and dispersion derivatives were not included. To be clear, I am not suggesting the authors reprocess their data an additional time using a different HRF selection (though I would consider it to be one way to resolve this criticism). However, in the absence of that step, I do believe that this should be discussed as an additional limitation (in the Discussion section) in reasonable detail for the rationale expounded above.

N.B. My apologies to the authors, in looking at my previous review, the citations I mentioned in comment 2-4 were cut off from the end of the review, and so I have ensured that they now appear in this version.

Lindquist, M. A., Loh, J. M., Atlas, L. Y., & Wager, T. D. (2009). Modeling the hemodynamic response function in fMRI: efficiency, bias and mis-modeling. NeuroImage, 45(1), S187-S198.

West, K. L., Zuppichini, M. D., Turner, M. P., Sivakolundu, D. K., Zhao, Y., Abdelkarim, D., ... & Rypma, B. (2019). BOLD hemodynamic response function changes significantly with healthy aging. NeuroImage, 188, 198-207.

• In the Introduction, the beginning the new third paragraph (beginning “The difficulty of seeing brief…”) ostensibly frames the expectation that neural responses in aged primary visual cortex, by way of reduced information processing capability, actually decrease with increasing temporal frequency. This is later confirmed explicitly in the next paragraph. While I believe this to be a reasonable and logical hypothesis based on what is presented in the Introduction, I was surprised to see no Discussion real estate whatsoever expended to explain the fact that the study finds the complete opposite. If the hypothesis was that BOLD signal will decrease monotonically with increasing checkerboard frequency, what would explain the observation that the opposite is actually true? I believe the authors should discuss this as best they can based on the analysis and methodology they used (and in the context of other work - I understand and applaud not wanting to speculate beyond data they observed).

• The authors have expanded on the rationale for utilizing surface-based analysis methods over volumetric methods. I agree with this rationale, and would suggest citing more literature (than just Brodoehl et al., 2020) to have an empirical basis upon which to base the expectation that surface-based methods will be superior to volumetric methods used in prior work. Specifically, I would like to see the citations below included.

Hutchison, J. L., Hubbard, N. A., Brigante, R. M., Turner, M., Sandoval, T. I., Hillis, G. A. J., ... & Rypma, B. (2014). The efficiency of fMRI region of interest analysis methods for detecting group differences. Journal of neuroscience methods, 226, 57-65.

Tucholka, A., Fritsch, V., Poline, J. B., & Thirion, B. (2012). An empirical comparison of surface-based and volume-based group studies in neuroimaging. NeuroImage, 63(3), 1443-1453.

• I found the description of the Visual processing speed/UFOV task to be lacking. The authors cited a 2017 article which does describe the task very well, which is fine, but I couldn’t quite discern what the second sentence was saying in the context of this manuscript. I don’t expect the authors to describe the task to the same extent as the 2017 paper, but a little more elaboration/clarification would be appreciated here.

• Finding negative BOLD signal relationships with the offset contrast (i.e., the general/frequency-independent effect of stimulus) with no such negative relationships with the slope contrast was interesting. The manuscript indicates that it was in line with previous studies, and cites [43-46], but no such entries appear in the References section. Please add these. Beyond that, though, I felt the (sixth) Discussion paragraph that asserts this as evidence for an experiment and analyses that were well-conducted was lacking. I feel that readers may not follow exactly why this result reinforces the experimental and analytical appropriateness. Please elaborate here (e.g., spatial nature of V1? Anterior/posterior?). I would also like to echo an earlier point – the finding of absolutely no significant clusters that showed a negative relationship to the slope contrast goes completely against the hypothesis laid out in the Introduction, so substantive discussion of one or more potential reasons is warranted.

Minor Point: Table 3 could use a little cleaning up visually (e.g., the word “Positive” is misspelled, the cells are only merged across columns within rows but gray row dividers are still visible).

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Jenny Rieck

Reviewer #2: Yes: Monroe P Turner

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

10 Sep 2021

Please find the attached file named "Response To Reviewers".

Submitted filename: Response_to_Reviewers20210909.docx

Click here for additional data file.

18 Oct 2021

BOLD signal response in primary visual cortex to flickering checkerboard increases with stimulus temporal frequency in older adults

PONE-D-21-00897R2

Dear Dr. Uchiyama,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Andrea Antal, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Jenny Rieck

Reviewer #2: Yes: Monroe P Turner